[gmx-users] Memory issues using -ac option of gmx hbond

2017-08-07 Thread Jared Sagendorf
I'm trying to analyzing hydrogen bonds between a DNA and protein complex.
When using the -ac option of gmx hbond, I'm getting segmentation faults
which look like an out-of-memory problem.

However, I'm giving the process 64gb of memory, and an additional 64gb of
virtual memory. How memory intenstive is this program??

Note I am using version 5.1.13. A snippet of the output is below

Calculating hydrogen bonds between DNA_interface (919 atoms) and
Protein_interface (1026 atoms)
Found 138 donors and 458 acceptors
Making hbmap structure...done.
...
Found 117 different hydrogen bonds in trajectory
Found 156 different atom-pairs within hydrogen bonding distance
Merging hbonds with Acceptor and Donor swapped
- Reduced number of hbonds from 117 to 116
- Reduced number of distances from 156 to 156
Average number of hbonds per timeframe 16.544 out of 31602 possible

 PLEASE READ AND CITE THE FOLLOWING REFERENCE 
D. van der Spoel, P. J. van Maaren, P. Larsson and N. Timneanu
Thermodynamics of hydrogen bonding in hydrophilic and hydrophobic media
J. Phys. Chem. B 110 (2006) pp. 4393-4398
  --- Thank You ---  

Doing autocorrelation according to the theory of Luzar and Chandler.
[hpc3338:34436] *** Process received signal ***
[hpc3338:34436] Signal: Segmentation fault (11)
[hpc3338:34436] Signal code: Address not mapped (1)
[hpc3338:34436] Failing at address: 0xfde7c370
[hpc3338:34436] [ 0] /lib64/libpthread.so.0(+0xf370)[0x7ff5c3810370]
[hpc3338:34436] [ 1]
/usr/usc/gromacs/5.1.3/cpu/lib64/libgromacs_mpi.so.1(+0x464449)[0x7ff5c4a7b449]
[hpc3338:34436] [ 2]
/usr/usc/gromacs/5.1.3/cpu/lib64/libgromacs_mpi.so.1(cross_corr+0x39)[0x7ff5c4a7b019]
[hpc3338:34436] [ 3]
/usr/usc/gromacs/5.1.3/cpu/lib64/libgromacs_mpi.so.1(+0x339bb2)[0x7ff5c4950bb2]
[hpc3338:34436] [ 4]
/usr/usc/gromacs/5.1.3/cpu/lib64/libgromacs_mpi.so.1(gmx_hbond+0x312e)[0x7ff5c4947aae]
[hpc3338:34436] [ 5]
/usr/usc/gromacs/5.1.3/cpu/lib64/libgromacs_mpi.so.1(_ZN3gmx24CommandLineModuleManager3runEiPPc+0x267)[0x7ff5c47b4637]
[hpc3338:34436] [ 6] gmx_mpi(main+0xba)[0x40bfea]
[hpc3338:34436] [ 7]
/lib64/libc.so.6(__libc_start_main+0xf5)[0x7ff5c2c01b35]
[hpc3338:34436] [ 8] gmx_mpi[0x40be69]
[hpc3338:34436] *** End of error message ***


Am I doing something wrong here?
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.


[gmx-users] gmx complains of mismatch between index and trajectory file

2017-07-27 Thread Jared Sagendorf
Hello, I'm using GROMACS 5.1.13 and ran into a problem I've never
encountered before. I've run a simulation on a zinc-finger DNA complex and
defined a group in my index filed called "Protein_DNA" from the default
"Protein" and "DNA" groups. This group should contain 4666 atoms, which it
does.

$gmx make_ndx -n index.ndx
0 System  : 180141 atoms
1 Protein :  2830 atoms <==
2 Protein-H   :  1427 atoms
3 C-alpha :   171 atoms
4 Backbone:   513 atoms
5 MainChain   :   685 atoms
6 MainChain+Cb:   851 atoms
7 MainChain+H :   854 atoms
8 SideChain   :  1976 atoms
9 SideChain-H :   742 atoms
10 Prot-Masses :  2830 atoms
11 non-Protein : 177311 atoms
12 Ion :32 atoms
13 ZN  : 7 atoms
14 NA  :25 atoms
15 DNA :  1836 atoms <=
16 ZN  : 7 atoms
17 NA  :25 atoms
18 Water   : 175443 atoms
19 SOL : 175443 atoms
20 non-Water   :  4698 atoms
21 Water_and_ions  : 175475 atoms
22 Protein_DNA :  4666 atoms <=

I'm not sure why the ZN and NA groups show up twice, but I assume it's not
an issue. Anyway, Protein_DNA is defined, and when I compare the values in
the index.ndx file to the atom numbers in the .gro file, it all matches up.

In my .mdp input file, I specify this group to be output as a compressed
.xtc trajectory.

; OUTPUT CONTROL OPTIONS =
nstxout  = 1
nstvout  = 1
nstfout  = 0
nstlog   = 1
nstenergy= 2500
nstxout-compressed   = 1000
compressed-x-precision   = 1000
compressed-x-grps= Protein_DNA

And then I run my simulation. I now have my traj.xtc file, which I want to
process. However, when using gmx trjconv, and select the group Protein_DNA,
I get the following error:
$gmx trjconv -f traj.xtc -n index.ndx [options]
Fatal error:
Index[4659] 4667 is larger than the number of atoms in the
trajectory file (4666). There is a mismatch in the contents
of your -f, -s and/or -n files.

The group DNA_Protein should contain 4666 atoms, and I've verified that my
traj.xtc file indeed contains 4666 items with gmx check:

$gmx check -f traj.xtc
Checking file traj.xtc
Reading frame   0 time0.000
# Atoms  4666
...

So what is the problem here? The error message states that "Index[4659]
4667 is larger than the number of atoms in the trajectory file (4666)",
which is true - the 4659th atom listed in the "Protein_DNA" index is indeed
atom 4667, but so what? There's 4666 atoms in the index, and 4666 atoms in
traj.xtc file - I don't see where the mismatch is.

Can someone help me out with this and suggest some things to try, or
explain what is wrong if I'm missing something?

Is it possible this is related to the fact that the protein topology file
includes 7 zinc atoms (since this is a ZF protein), which are not included
in the group Protein_DNA? Note that if I don't supply an index file and
choose "Protein" from the default groups, gmx
trjconv will run, but if I choose "DNA", I get a similar error.

Fatal error:
Index[1829] 4667 is larger than the number of atoms in the
trajectory file (4666). There is a mismatch in the contents
of your -f, -s and/or -n files.
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.